Skip to main content
NCBI home page
SSM - Population Health logoLink to SSM - Population Health
. 2018 Jul 21;5:210–217. doi: 10.1016/j.ssmph.2018.07.002

Weighed down by discriminatory policing: Perceived unfair treatment and black-white disparities in waist circumference

Michael J McFarland a,⁎, John Taylor a, Cheryl AS McFarland b
PMCID: PMC6072653  PMID: 30094316

Abstract

Police maltreatment, whether experienced personally or indirectly through one’s family or friends, represents a potentially harmful stressor, particularly for minority populations. We address this issue by investigating: (1) how waist circumference (WC) varies by personal and vicarious exposure to unfair treatment by police (UTBP); and (2) to what extent exposure to UTBP explains the black-white disparity in WC. We employed data collected from a community-based sample of black (n = 601) and white (n = 608) adults living in Nashville-Davidson county Tennessee to address these questions. Results from our final linear regression model showed that those who reported vicarious UTBP had WCs that were approximately 2 in. greater than those who did not (b = 2.03; p = 0.003). While personal UTBP was not linked to higher WC, a post-hoc analysis suggested that our ability to detect an association was complicated by selection. Binary mediation analysis revealed that differential exposure to vicarious UTBP accounted for approximately 12% of the black-white WC disparity among women. We found no black-white differences in WC among men. The association between vicarious UTBP and WC did not vary by age, race, or gender. Overall, our findings point toward the role of discriminatory policing as a potential upstream contributor to racial disparities in health.

Keywords: Race, Police, Discrimination, Obesity, Waist circumference, Adiposity

Highlights

  • •

    Unfair treat by police (UTBP), whether experienced personally or vicariously, is a stressor that disproportionately affects minority populations

  • •

    Exposure to vicarious UTBP is linked with higher waist circumferences (WCs)

  • •

    Black women have larger WCs than white women

  • •

    Exposure to vicarious UTBP explains 12% of the black-white disparity in WC among women

1. Introduction

1.1. Policing, race, and health

Recent high-profile killings of unarmed men by police were pivotal in the emergence of The Black Lives Matter movement (BLM). The BLM, however, points to these slayings as the visible surface of a much wider and deeply-rooted problem where police regularly intrude on the lives of black citizens (Garza, 2014). A new wave of research seems to corroborate these claims by providing evidence that discriminatory policing represents a potential upstream detriment to population health. For instance, a small but growing body of research shows that aggressive policing is linked to poor mental health (Oh et al., 2017, Geller et al., 2014, DeVylder et al., 2017a, DeVylder et al., 2017b) and high levels of police surveillance - the regular monitoring and stopping and search citizens – is linked to increased psychological distress, obesity, and hypertension (Sewell and Jefferson, 2016, Sewell et al., 2016, Sewell, 2017).

Also consistent with the claims of the BLM, exposure to police encounters occurs unevenly throughout society with the most invasive policing occurring among minority populations (Fagan et al., 2010, Bjornstrom, 2015, Glaser, 2015). Nationally representative surveys of US adults show that approximately 40–60% of blacks but only 7–10% of whites report ever having been treated unfairly by police (e.g. Weitzer & Tuch, 2005; Robert Wood Johnson 2017). Community-level data on police stops and searches also point to the heavy burden faced by minority populations. Data collected by the Metro Nashville Police Department (MNPD) and analyzed by a nonprofit organization shows that traffic stops in Nashville predominantly occur in low income, black, or Hispanic neighborhoods and that black drivers are five times more likely to be stopped multiple times per year compared to white ones. Moreover, blacks are searched during automobile stops at a rate more than two times that of whites despite the fact that incriminating evidence is found at a higher rate among whites (Gideon’s Army, 2016).

In this study we posit that UTBP, experienced personally or vicariously through a close friend or family member, represents a virulent stressor contributing to racial disparities in obesity. UTBP is defined as interactions in which an individual perceives that police are deviating from procedural justice– unequally enforcing the law, making disrespectful comments or gestures, making unnecessary stops and searches, or making threats (Tyler & Wakslak, 2004). Our focus on vicarious exposure, represents a unique strength of this study. While exposure to vicarious racism has received a fair amount of scholarly attention (e.g. Heard-Garris, Cale, Camaj, Hamati, & Dominguez, 2017), vicarious exposure to UTBP, in particular, has received very little. However, vicarious exposure to UTBP may be particularly important within the context of stress and health. Black parents often worry that their children will encounter a life shattering interaction with police. This fear has led to the culturally ubiquitous “talk” where black parents instruct their children on how to safely navigate police encounters (Thomas, 2013, Whitaker and Snell, 2016). Given the scope of these phenomena it seems plausible and perhaps likely that having a family or close friend experience UTBP will have detrimental health consequences.

1.2. Unfair treatment by police as a stressor

UTBP is conceptualized as a discrete event that is transformed into a chronic strain because it precipitates ruminative stress and heightened vigilance. When UTBP occurs, the inherent power differential between the police and the public is particularly meaningful. Police are afforded tremendous power to use physical violence, impose legal sanctions, and engage in other forms of coercion and inducements by the state. UTBP simultaneously represents a threat to physical safety, freedom, financial wellbeing, sense of agency, and sense of self (Harris, 2010, Brunson, 2007, Henning, 2017). One study, for instance, showed that exposure to racial discrimination was associated with increased body mass index (BMI) and waist circumference because those who experienced discrimination also experienced chronic vigilance (Hicken, Lee, & Hing, 2017). This pattern may extend to exposure to UTBP as well.

Criminological research shows that vicariously experienced instances of police maltreatment can result in enduring worry and the erosion of trust in law enforcement (e.g. Warren, 2011; Browning, Cullen, Cao, & Kopache, 1994). Moreover, research suggests that negative vicarious experiences alter perceptions of police to a greater extent than interpersonal encounters (Rosenbaum, Schuck, Costello, Hawkins, & Ring, 2005). Given the pronounced link between exposure to vicarious racism, generally, and worse health (e.g. Heard-Garris et al., 2017; Priest et al., 2013; Williams & Mohammed, 2013), it seems plausible that vicarious exposure to UTBP may also be linked to worse health. In particular, having a family member or close friend experience UTBP may impact health in two possible ways. First, vicarious experiences may initiate or intensify worry about one’s friends and family’s future encounters with police and the ramifications of these interactions. Second, these encounters may induce worry and chronic vigilance that similar encounters may happen to them personally. For instance, knowing someone who was victimized often leads to fear that one will be personally victimized (Agnew, 2002).

1.3. Waist circumference and the stress and health latency problem

We focus on obesity, as measured by waist circumference (WC), specifically because we are able to partially overcome the stress and illness latency problem. When stressors are experienced at younger ages they can take decades to manifest as chronic health conditions. WC, in contrast, can change in response to stress in much shorter time spans (i.e. weeks, months, or years rather than decades). Exposure to stress is known to elicit eating foods that are high in calories and saturated fats which, in turn, produce a biochemical response that reduces feelings of distress (Dallman, Pecoraro, & la Fleur, 2005). Moreover, stress is also thought to increase one’s chances of being obese through dysregulation of the hypothalamus-pituitary adrenal (HPA) axis which can result in appetite dysregulation, decreased signaling of satiety, and the accumulation of visceral adiposity (Adam & Epel, 2007).

1.4. Black-White disparities in waist circumference

Black-white disparities in obesity are well documented with blacks being 2–3 times more likely to be obese and have WCs that are considerably greater than their white counterparts (Ogden et al., 2014, Krueger and Reither, 2015, Wang and Beydoun, 2007). The leading hypothesis thought to explain this difference, termed the differential exposure hypothesis, suggests that these disparities emerge and persist because blacks are faced with a higher stress burden than whites (Hummer, 1996; Adler, Boyce, Chesney, Folkman, & Syme, 1993). While past work partially supports this hypothesis, racial variation in exposure to commonly measured social stressors only accounts for a small proportion of racial differences in health outcomes, including obesity (e.g. Sternthal, Slopen, & Williams, 2011). It therefore seems plausible that researchers are failing to consider other important racially patterned stressors (Turner, 2012, Williams and Mohammed, 2013). We advance UTBP as one such stressor that has not been integrated well into health disparities research.

1.5. Gender and racial disparities in waist circumference

Racial disparities in obesity-related outcomes including BMI and waist circumference are more pronounced among females (Jackson et al., 2013, Mensah et al., 2005, Wang and Beydoun, 2007). Moreover, some research reports that exposure to stressors is more likely to result in increased BMI among females than males (Hicken, Lee, & Hing, 2017). In this study, we test whether UTBP exposure partially accounts for black-white disparities in WC separately for women and men.

1.6. Research aims and hypotheses

Based on these arguments, we present three hypotheses. First and second, those who report personally or vicariously experiencing UTBP will have higher WCs that those who do not. Third, the black-white disparity in WC will be partially explained by blacks’ disproportionate exposure to UTBP. This hypothesis is presented in the context of knowing that black-white disparities in obesity and WC are usually more pronounced among women. In addition, we perform exploratory analyses regarding potential gender, race, and age-based heterogeneity in the relationship between UTBP and WC. We test these hypotheses using a community-based sample of black and white adults (ages 22–69) living in Nashville-Davidson county Tennessee.

2. Data and methods

2.1. The setting

Nashville represents an ideal setting to evaluate this study’s aims because it is a large diverse southern urban city with evidence of simmering racial tensions between police and communities of color. For example, MNPD’s controversial Operation Safer Streets has been publicly accused of criminalizing minorities (Gonzalez, 2016). Nashville also represents an ideal setting because the MNPD engages in proactive policing – practices that include actively engaging citizens through stops and searches in high-crime areas to detect expectant criminal activity in order to disrupt situations interpreted to be indicative of crime (Geller et al., 2014). Moreover, interviews with black community members described police frequently engaging in aggressive behavior and making racial insults and threats of violence (Gideon’s Army, 2016). Overall, if UTBP exposure is linked to racial differences in WC, Nashville represents a city where we are likely to detect this phenomenon.

2.2. Study sample

Data come from the Nashville Stress and Health Study (NSAHS), a probability sample of non-Hispanic black and white adults living in Davidson County, TN (http://vanderbilt.edu/stressandhealthstudy/). Survey Sampling International generated a random sample of 199 block groups in Davidson County. To adequately sample black households, block groups were stratified by the percentage of black residents (2010 Census). The study design included a random sample of 1252 adults living in the sampled and stratified block groups. Approximately 61% of contacted households agreed to participate in the study. The average computer-assisted interview lasted approximately 3 h. Trained same-race interviewers conducted all interviews in the respondent’s home or on Vanderbilt University’s campus.

The primary purpose of NSAHS is to examine black/white and socioeconomic disparities in health. The morning after the in-person interview, a trained clinician visited the respondent to measure height, weight, waist circumference, and other metabolic indicators. Data collection occurred between April of 2011 and January of 2014. We handled missing data on UTBP (n = 1), parental education (n = 112), and financial strain in childhood (n = 1) through a regression-based imputation procedure (Landerman, Land, & Pieper, 1997). Forty-three cases with missing WC data were dropped giving a final sample size of N = 1,209.

2.3. Key measures

Waist circumference was measured by trained clinical interviewers using standard protocols put in place by the Allostatic Load Working Group of the McArthur Research Network. WC was measured in inches using a tape measure just above the upper hip bone. All analyses were replicated using BMI as an alternative measure of adiposity.

Perceived unfair treatment by police comes from one question from the major discrimination measure (Williams, Yu, & Jacksonm 1997). Respondents were asked “Have you or anyone close to you ever had the following experience: ”been treated unfairly by police (e.g., stopped, searched, questioned, physically threatened or abused)?” Those responding “yes” were probed whether it happened to them, a spouse, child, or another relative or close friend. Those answering affirmatively to experiencing personal UTBP led to routing past the vicarious UTBP response. This prohibited us from identifying those that experienced both personal and vicarious UTBP. Two variables were created that indicated whether the respondent personally experienced UTBP or vicariously experienced it (Yes = 1). Those identified as having experienced vicarious UTBP did not experience personal UTBP.

We also utilized an item that asks the respondent about when UTBP occurred. The interviewer asked “when was the last time this occurred?” Responses including 1 “in the last month”, 2 “in the last six months”, 3 “in the last year”, and 4 “over a year ago.”

2.4. Statistical controls

This study includes an extensive suite of covariates to address potential spuriousness. We identified the following factors as important controls because they are feasibly linked to race, UTBP, and WC: sociodemographic characteristics, stressful life events, and exposure to other types of discrimination, (Geller et al., 2014, DeVylder et al., 2016, Williams and Mohammed, 2013; Burdette & Hill, 2008). Mental health disorders are also considered because they are frequently linked to obesity and police disproportionately interact with those that have higher rates of mental health disorders (Sugie & Turney, 2017). Similarly, we included indicators of previous trouble with the law controls because UTBP may be correlated with any type of contact and WC (Archibald, Parker, & Thorpe, 2018).

Sociodemographic characteristics were measured with several variables including: age (continuous years), educational attainment of parent (college degree = 1), and financial strain in childhood. The latter variable asked about their family’s ability to pay for basic necessities with answers ranging from 1 “could easily afford” to 5 “could not afford.” Years of education measured years of formal education completed. Financial strain in adulthood was measured as a mean response to six items asking about difficulty in meeting basic needs (Chronbach’s α = 0.82). Stressors were measured using two variables including: the number of stressful life events prior to age eighteen and the number of stressful life events experienced after age seventeen (see McFarland, Taylor, Hill, & Friedman, 2018).

Discrimination was measured by two indices developed by Williams, Yu, Jackson, and Anderson (1997). Perceived everyday discrimination was based on seven items that asked the respondent about the frequency they experience discrimination (e.g. treated with less courtesy than other people, called names). Responses ranged from never “1” to almost always “5” and were averaged (Chronbach’s α = 0.85). Our measure of perceived unfair treatment asked if the respondent or someone close to the respondent had experienced unfair treatment in various life domains (e.g. school, work, and housing). We summed the number of domains in which unfair treatment occurred (see Williams et al., 1997). UTBP was excluded from this count variable. Mental health was measured with two variables. Ever having had generalized anxiety (yes = 1) or major depressive disorder (yes = 1) were measured via the Composite International Diagnostic Interviewer (CIDI) (for more information see Turner & Lloyd, 1999).

Legal problems were measured using three dichotomous indicators: the respondent was stopped or arrested for alcohol more than once (yes=1), was accused or arrested of crime in last 12 months (yes=1), and had someone close accused or arrested of crime in last 12 months (yes=1).

2.5. Analytical strategy

We begin our analysis by presenting descriptive statistics stratified by race and conduct bivariate tests of whether WC varies by race (see Table 1). Descriptive statistics are also stratified by gender because prior evidence suggests the black-white disparity in WC might be limited to women. Whether this is the case in our data is important because it will tell us whether to include men in the mediation analysis. We subsequently present descriptive statistics stratified by gender and UTBP (see Table 2) and conduct bivariate tests of whether WC varies by UTBP. Next, we turn our attention to how UTBP is linked to WC after accounting for statistical controls in Ordinary Least Squares (OLS) multivariate regression models (Table 3). Our general strategy is to first test for associations using a baseline model with minimal demographic control variables and then test for these associations after including all covariates. Model 1 tests for black-white disparities in WC. Model 2 includes measure of UTBP into the model and tests for associations between UTBP and WC. Models 3 and 4 replicate those of 1 and 2 except they include the full set of statistical controls. We also show these models stratified by gender. Multicollinearity did not pose a problem, as the highest variance inflation factor did not exceed 2.02.

Table 1.

Descriptive statistics stratified by race and gender (N = 1, 209).

Blacksd Whitese Black Womenf White Womeng Black Menh White Meni Range
(n = 601) (n = 608) (n = 315) (n = 323) (n = 286) (n = 285)
Mean/Percent Mean/Percent Mean/Percent Mean/Percent Mean/Percent Mean/Percent Min Max
Waist circumference (inches) 40.05*** 37.94*** 41.06*** 36.74*** 38.94 39.30 23.5 126
Personal unfair treatment by police (% yes) 24.63*** 10.86*** 13.97*** 6.19*** 36.36*** 16.14*** 0 100
Vicarious unfair treatment by police experienced by family/friend (% yes) 21.13*** 7.73*** 26.35*** 8.67*** 15.38*** 6.67*** 0 100
Age 46.01 45.84 45.33 46.44 46.76 45.16 22 69
Female (% female) 52.41 53.13 – – – – 0 100
Parent has college degree (% yes) 16.14*** 32.24*** 17.46*** 29.41*** 14.69*** 35.44*** 0 100
Financial strain in childhood 2.76*** 2.41*** 2.75*** 2.42*** 2.77*** 2.40*** 1 5
Years of education completed 13.05*** 14.99*** 13.17*** 15.00*** 12.92*** 14.98*** 1 25
Current financial strain 1.58*** 1.42*** 1.61*** 1.46*** 1.56*** 1.38*** 1 3
Stressful life events experienced in childhood (count) 3.57** 3.06** 3.42 3.00 3.73* 3.14* 0 18
Stressful life events experienced in adulthood (count) 3.69*** 2.99*** 3.89*** 3.14*** 3.47** 2.83** 0 13.2
Discrimination 2.11*** 1.92*** 2.09*** 1.90*** 2.13*** 1.95*** 1 4.78
Perceived unfair treatment (count) 1.46*** 1.09*** 1.51*** 1.15*** 1.41*** 1.03*** 0 6
Anxiety disorder (% ever) 3.83*** 10.03*** 5.71*** 13.62*** 1.75* 5.96* 0 100
Depression (% ever) 12.81*** 30.26*** 16.83*** 34.06*** 8.39*** 25.96*** 0 100
Stopped/arrested for alcohol more than once (% yes) 3.99 2.47 1.59 2.17 6.64* 2.81* 0 100
Accused/arrested of crime in last 12 months (% yes) 3.00 2.80 1.27 2.48 4.90 3.16 0 100
Family/friend accused of crime in last 12 months (% yes) 11.48 9.05 13.33 10.84 9.44 7.02 0 100
Open in a new tab
*

p<0.05 based on a two-sample t-test or a Pearson χ² test by race.

**

p<0.01,

***

p<0.001,

d

compared to whites.

e

compared to blacks.

f

compared to white women.

g

compared to black women.

h

compared to white men.

i

compared to black men.

Table 2.

Descriptive statistics stratified by gender and perceived unfair treatment by police.

Women
 Men
Unfair to Self (n = 64) Unfair to Other (n=111) No Unfair Treatment (n = 463) Unfair to Self (n = 150) Unfair to Other (n = 63) No Unfair Treatment (n = 358)
Waist circumference (inches) 39.54 41.68** 38.10 39.08 40.37 38.92
Age 44.78 44.39 46.41 44.22* 45.46 46.78
Parent has college degree (% yes) 14.06* 23.42 24.89 23.33 25.81 25.63
Financial strain in childhood 2.73 2.65 2.55 2.74** 2.78* 2.49
Years of education completed 13.09** 13.81 14.30 12.98*** 14.05 14.33
Current financial strain 1.84** 1.53 1.49 1.59*** 1.50 1.41
Stressful life events experienced in childhood (count) 5.19*** 3.67* 2.82 4.97*** 3.41 2.79
Stressful life events experienced in adulthood (count) 5.24*** 4.31*** 3.08 3.45** 4.21*** 2.84
Discrimination 2.50*** 2.11*** 1.89 2.25*** 2.19*** 1.93
Perceived unfair treatment count (count) 1.92*** 1.96*** 1.09 1.76*** 1.75*** 0.90
Anxiety disorder (% ever) 9.38 14.41 8.64 2.00 3.77 4.75
Depression (% ever) 32.81 24.32 24.84 14.67 17.46 18.16
Stopped/arrested for alcohol more than once (% yes) 7.81** 2.70 0.86 11.33*** 1.59 2.51
Accused/arrested of crime in last 12 months (% yes) 7.81*** 0.90 1.30 9.33*** 1.59 2.23
Family/friend accused of crime in last 12 months (% yes) 18.75 * 18.92 ** 9.50 10.00 15.87 ** 6.15
Open in a new tab

P values were assessed based off the results of either a two-sample t-test or a Pearson χ2 test.

*

p < .05;

**

p < .01;

***

p < .001 based on a within-race test compared to those not reporting unfair treatment.

Table 3.

Waist circumference regressed on race and unfair treatment by police.

Full Sample (N = 1209)
 Women (N = 638)
 Men (N = 571)
Model 1e Model 2e Model 3f Model 4f Model 5e Model 6e Model 7f Model 8f Model 9e Model 10e Model 11f Model 12f
Black (ref. = whites) 1.40* 0.99 1.44* 1.16a 2.66** 2.17* 3.04** 2.68** -0.21 -0.45 -0.34 -0.53
(0.60) (0.61) (0.62) (0.63) (0.91) (0.92) (0.95) (0.97) (0.80) (0.81) (0.82) (0.83)


 

 

 

 

 

 

 

 

 

 

 

 


Personal UTBPg – 0.01 – -0.17 – 0.32 – -0.64 – 0.56 – 0.47
– (0.62) – (0.66) – (1.15) – (1.24) – (0.73) – (0.80)


 

 

 

 

 

 

 

 

 

 

 

 


Vicarious UTBPg – 2.36*** – 2.00** – 2.37* – 2.04* – 1.86a – 1.74a
– (0.61) – (0.68) – (0.94) – (0.97) – (0.95) – (1.00)
Adjusted R2 0.05 0.06 0.06 0.13 0.14 0.13 0.14 0.00 0.01 0.00
Open in a new tab
a

p <.10;

*

p <. 05;

**

p <. 01;

***

p< .001 (two-tailed tests); standard errors in parentheses. UTBP = unfair treatment by police.

e

These models include covariates for age, gender, parental education, financial strain in childhood, and fixed effects for census tract.

f

Includes all study covariates.

g

The reference group consists of those that did not report any type of unfair treatment by police.

Next, we conducted mediation analysis to test the extent to which UTBP mediates the black-white disparity in WC using binary mediation models. Because we only find a black-white disparity in WC among women, we restrict our mediation analysis to women as well. Our two proposed mediators are dichotomous, thus the path coefficients and standard errors were rescaled (see http://www.nrhpsychcom/mediation/logmed.html) prior to calculating indirect associations (not shown). Mediation analyses were conducted using STATA’s binary mediation command (see Ender, 2011) and standard errors were calculated using bootstrapping methodology with 1000 replications. All regression models included fixed effects for census tracts because UTBP may be a proxy for living in heavily policed neighborhoods with high levels of crime. These fixed effects help guard against omitted variable bias due to neighborhood conditions. Finally, we tested whether any associations between UTBP (both personal and vicarious) and WC varied by race, age, and race and gender. They did not (see Appendix B).

3. Results

3.1. Unfair treatment by police and waist circumference

Table 1 presents black-white compositional differences in means or percentages for all study variables. WC was normally distributed (meanblacks = 40.07, standard deviation = 7.84; meanwhites = 37.93, standard deviation = 6.97) with blacks having higher levels than whites (t = 5.0; p < .001). Table 1 also shows that this disparity only exists among women. Perceived UTBP was fairly common in this sample with 32% of the full sample reporting either personal or vicarious unfair treatment. Reports of UTBP differed markedly by race, however. Blacks were more likely to report personal (24.9%) and vicarious (20.9%) UTBP compared to their white counterparts (personal UTBP: 10.8%; χ2 = 41.1; p < 0.001; vicarious UTBP: 7.7%; χ2 = 43.1; p < .001). Overall, 45.8% of blacks reported personal or vicarious UTBP compared to 18.5% among whites (χ2 = 103.3; p < 0.001). Men were more likely to report personal UTBP while women were more likely to report vicarious UTBP. Blacks were also more disadvantaged on socioeconomic indicators and had more exposure to stressors and discrimination. Whites were more likely to have had depression or generalized anxiety disorder.

Table 2 presents descriptive statistics stratified by gender and UTBP. Women who reported personal UTBP did not differ in WC from women who reported no UTBP. Women who experienced vicarious UTBP had higher WCs than women who did not report UTBP (p ˂ 0.01). Men who reported personal or vicarious UTBP did not differ in WC from those who did not report UTBP.

Pronounced within-race compositional differences were observed depending on exposure to personal, vicarious, or no exposure to UTBP for blacks and whites (not shown). In particular, blacks who experienced personal UTBP tended to be disadvantaged compared to their non-exposed counterparts on financial strain, stressful life events, discrimination, unfair treatment, and past legal history. Blacks who experienced vicarious UTBP were fairly similar to those who did not experience UTBP but were disadvantaged in terms of stressful life events in adulthood, discrimination, and unfair treatment. These compositional differences were similar among whites.

Pronounced within-gender compositional differences were observed as well as seen in Table 2. In particular, women who experienced personal UTBP tended to be disadvantaged compared to their non-exposed counterparts on exposure to financial strain, educational attainment, stressful life events, discrimination, unfair treatment, and past legal history. Women who experienced vicarious UTBP were disadvantaged in terms of exposure to stressful life events in adulthood, discrimination, unfair treatment, and having a family member or friend accused of a crime. These compositional differences were similar among men. Overall, Table 2 presents support that vicarious UTBP exposure is linked to higher WCs, but pronounced compositional differences that vary by UTBP exposure suggest multivariate test of these associations are required.

Table 3 presents the results of a series of multivariate models predicting WC. Model 1 reveals marked racial disparities in WC as blacks had elevated WC compared to whites (b = 1.42; p <.05). Model 2 shows that personal UTBP was unrelated to WC while those who experienced vicarious UTBP had higher WCs compared to those who did not experience UTBP (b = 2.37; p < 0.001) after controlling for baseline demographic characteristics. The inclusions of UTBP into model 2 decreased the black-white difference in WC (i.e. from 1.42 to 1.01). Models 3 and 4 show that these same results remain after including the full array of study controls. In particular, vicarious UTBP was positively associated with WC (b = 2.02; p <0.01) and the inclusion of this variable into the model decreased the black-white disparity in WC (i.e. from 1.40 to 1.16). Table 3 also shows the black-white disparity in WC is only extant among women, personal UTBP was not associated with WC among women or men, and vicarious UTBP was positively associated with WC among both women and men. Overall, Table 3 provides suggestive evidence that vicarious UTBP may mediate the black-white difference in WC seen among women.

3.2. Probing for evidence of selection

Because we found no evidence for our hypothesis that personal UTBP and WC would share a positive association (see Table 3), we performed a post hoc analysis to test whether those who experienced personal UTBP had lower WC’s at the time of exposure − which would provide evidence for health selection. If this is the case, we would expect to see people with the smallest WC near the time of exposure and observe increases in WC with increases in time since exposure. We tested this possibility by regressing WC on time since UTBP, an ordinal variable described in the measures section. We used a linear regression model after including controls for race, gender, age, parental education, and financial strain in childhood and limiting the model to those who reported personal UTBP. The results of this regression were plotted in Fig. 1. It shows a positive association between time since personal UTBP and WC (b = 1.28; p = 0.06). The mean WC for those not exposed to any UTBP is also shown as a point of comparison in the figure. No evidence was found for selection among those reporting vicarious UTBP (b = -0.69; p = 0.33).

Fig. 1.

Fig. 1

Open in a new tab

Average Waist Circumference by Amount of Time. Since Last Experience of Personal Unfair Treatment by Police. Notes: The solid line reflects the positive association between time since personal UTBP and waist circumference based on a linear regression among those that reported a personal experience of unfair treatment by police (n= 214). Controls for age, gender, race, parental education, and financial strain in childhood were included. The dashed line signifies that those that did not report any type of UTBP were not included in the regression. The average waist circumference among those that did not report any type of unfair treatment by police is presented as a point of reference.

The inclusion of some older adults in the model may be conflating the association between personal UTBP and WC. In order to identify those that were likely to have experienced policing outside of the proactive policing model, we tested whether there was an association between age – collapsed into three categories 22–35, 36–50, and 51–69 – and time since personal UTBP. We found that those ages 51–69 were more likely to report UTBP that occurred more than a year ago (χ2 = 12.7; p < 0.05) than their younger counterparts. While we cannot definitively identify those who experienced UTBP prior to implementation of proactive policing practices, we identified this group as especially likely to have done so.

Next, we removed those who experienced personal UTBP in the last year and those who likely experienced personal UTBP in the distant past (i.e. those over fifty that experienced UTBP more than a year prior to the study) from our analysis and re-ran models 2 (for both men and women), 6 (women only), and 10 (men only) from Table 3. These models showed that women and men who reported personal UTBP had WCs that were 2.5 and 1 in. larger, respectively, than their non-exposed counterparts (See Appendix Table A). While the magnitude of these associations were consistent with those between vicarious UTBP and WC (bwomen and men = 1.02; p = 0.21; bwomen = 2.46; p = 0.12; bmen = 1.05; p = 0.22), they were not statistically different from zero. Limited power was a concern. Assuming the association between UTBP and WC was two, the power to detect these associations in models 1–3 was 0.74, 0.47, and 0.42, respectively.

3.3. Mediation analysis

Fig. 2 present our mediation analyses for UTBP among women. We find a significant indirect association for vicarious (z = 2.32; p< 0.05) but not personal (z = -0.86; p > 0.05) UTBP. Based on this analysis, 12% of the black-white disparity in WC was explained by vicarious UTBP. Taken together, our analyses suggest that black women have higher WCs than white women, in part because they are more likely to experience vicarious UTBP. There is no evidence to support personal UTBP as a mechanism underlying the black-white gap in WC.

Fig. 2.

Fig. 2

Open in a new tab

Mediation analysis predicting WC among women with all study control variables included as covariates. Reported coefficients are unstandardized. UTBP = Unfair treatment by police. Z statistics and standard errors were calculated using a bootstrap method with 1000 replications. Proportion of total association mediated due to UTPB experienced by family or friend = 0.12. ***p < .001, *p < .05 (two-tailed). Standardized indirect effect for personal UTBP = -.01. Standardized indirect for vicarious UTBP = 0.03.

3.4. Robustness checks

We tested the possibility of subgroup variation in the associations between UTBP and WC using two OLS regression models for eight subpopulations. Our samples were stratified by race (i.e. black and white), age (i.e. those for younger than thirty compared to those thirty or older), and both race and gender (i.e. black females, white females, black men, and white men). The first model tests for these associations net of basic controls while the second tests them after including all study controls. We tested whether the UTBP coefficients varied across subpopulation using a t-test. Overall, our results are consistent with our main results from Table 3 as none of the coefficients varied by subpopulation (See Appendix A). All study findings were replicated using BMI as the outcome. These analyses provided identical substantive results as those reported here. Finally, other ancillary models showed our key findings were robust in models that examined personal and vicarious UTBP separately.

4. Conclusion

This study showed that those reporting vicarious UTBP had higher WCs than those not reporting UTBP. Moreover, exposure to vicarious UTBP explained about 12% of the black-white disparity in WC among women. Personal UTBP was not associated with WC and did not contribute to the black-white gap in WC.

Our data point to two phenomena that may have obfuscated our efforts to detect an association between personal UTBP and WC. First, it seems plausible that the people most at risk for experiencing personal UTBP spend more time in public places and are slimmer than their counterparts than their less at-risk counterparts. Evidence shows that people who spend more time doing indoor activities, like watching television, using the computer, and working on arts and crafts, are more likely to be obese than their counterparts who spend more time on activities outside the home (Patel, Spaeth, & Basner, 2016). If personal UTBP leads to increased WC and this type of selection is occurring we would expect that WC should be larger as time since personal UTBP increases.

Second, older adults may have experienced a qualitatively different kind of UTBP because of historical shifts in policing style over the past 20–25 years (Heymann, 2000). In particular exposure to personal UTBP in the context of the proactive policing model may have been more detrimental than similar exposure outside of proactive policing practices because of worry and chronic vigilance about future encounters. To test whether these phenomena might be preventing us from detecting a true association, we excluded those who reported UTBP as occurring within the last year and those older than fifty from our analysis. We found suggestive evidence that those reporting personal UTBP had larger WCs than those that did not. This evidence was only suggestive because the key associations were not statistically significant – which was likely due limited statistical power.

Our findings suggest there are collateral consequences for people of color that have not been previously considered when evaluating the ramifications of proactive policing practices. Proactive models of policing advocate methods such as stopping people for minor crimes or no crimes at all. From a community level, the justification for such action is that it leads to the arrest and detainment of a small number of potentially violent criminals and thus decreases the amount of violent crime more broadly. The utility of these practices has been a subject of debate among researchers and policy makers with the general consensus that small decreases in violent crime are possible with hefty increases in proactive and invasive policing (e.g. Fagan, 2017). At a personal level, the stopping, searching, and detaining of innocent people has been argued by proactive policing advocates to be little more than a nuisance for innocent people (Kelling and Bratton, 2015, Kelling and Cole, 1996), implying that the costs of inconveniencing people are worth the benefits of decreasing violent crimes. Our research counters this argument by revealing the “stress burden” incurred by those who vicariously experienced UTBP.

Several study limitations are noteworthy. First, UTBP was measured retrospectively and is therefore susceptible to recall errors, especially age-based recall errors. While this limitation remains a concern, its seriousness as a threat to our findings is mitigated by the fact that recall bias generally leads to underreports of stressful life events (Glasner & Van der Vaart, 2009). Thus, our tests were likely conservative estimates of the association between UTBP and WC. Second, we were not able to measure one’s history of incarceration, which may have led to a spurious association. This limitation is offset, however, by the finding that UTBP was only related to WC among those who experienced it vicariously. Third, our external validity is limited to a small geographical area.

Our indicators of UTBP were quite blunt, did not account for one’s the life course history with police, and did not include other relevant factors that may imbue police interactions with a particular sense of salience and meaning. Future research should assess police interactions and pay particular attention to the frequency, severity, timing, and context in which they occur. The extent to which a stressor is perceived as stressful is dependent on its perceived threat, predictability, and ability to control it. The frequency, type, and severity of UTBP exposure will likely combine with one’s personal, network-based and collective experiences with police, to influence these characteristics. The current study does not have such exposures. Moreover, based on our measure of UTBP we do not know how far in the past (beyond one year) the event occurred. Finally, longitudinal data would have been optimal for mediation analysis and the utilization of cross-sectional data thus represents a limitation (MacKinnon, 2008). This limitation, however, is somewhat mitigated by the fact that reverse causation seems unlikely in the three associations underlying the mediating influence of vicarious UTBP. In particular, it seems unlikely that WC influenced one’s exposure to vicarious UTBP.

This study adds to a growing body of work pointing to discriminatory policing practices as a potential upstream contributor to population health and racialized health disparities (Geller et al., 2014, Sewell et al., 2016, Sewell, 2017; Sewell, 2016; DeVylder et al., 2016). Given that scholars are frequently unable to empirically account for the social factors that underlie racial disparities in health and mortality, UTBP represents a promising avenue of study that may help do so. This study points to the “stress burden” of vicarious UTBP incurred by African Americans in the Nashville area. To the extent that exposure to UTBP is shaped by top-down police policies, alterations to these policies might serve as a lever capable of improving the health of minorities living in urban areas. For this reason, future research that examines linkages between police practices and policies, discriminatory policing, and health disparities is urgently needed.

Ethics

This research uses de-identified human data that does not meet the definition of human subjects research.

Appendix Table A. Waist circumference regressed on unfair treatment by police among those fifty or younger that did not report unfair treatment within the last year

Full Sample (N = 628) Women (N = 337) Men (N = 291)
Model 1a Model 2a Model 3a
Personal UTBP 1.023 2.463 1.050
(0.805) (1.585) (0.853)
Vicarious UTBP 1.712 0.975 2.545 *
(0.889) (1.232) (1.234)
Adjusted R2 0.04 0.13 0.04
Open in a new tab

* p <. 05 (two-tailed tests); standard errors in parentheses. UTBP = unfair treatment by police.

a This model includes covariates for age, race, gender, parental education, and financial strain in childhood.

Appendix B. Waist circumference regressed on race and unfair treatment by police

Panel A
Model 1a
 Model 2b
 Model 3a
 Model 4b
 Model 5a
 Model 6b
 Model 7a
 Model 8b
Blacks (n = 601) Whites (n = 608) Age < 30 (n = 100) Age ≥ 30 (n = 1,109)
Personal UTBP 0.15 -0.11 -0.15 0.02 -4.47 -0.92 -0.15 -0.31
(0.86) (0.93) (0.95) (1.04) (4.60) (5.43) (0.64) (0.69)
Vicarious UTBP c 1.65† 1.29 3.16** 3.10 5.12 3.30 2.34** 1.99†
(0.87) (0.90) (1.08) (1.13) (3.43) (3.74) (0.69) (0.71)
Adjusted R2 0.02 0.02 0.17 0.17 -0.08 -0.06 0.04 0.05
Panel B
Model 9a
 Model 10b
 Model 11a
 Model 12b
 Model 13a
 Model 14b
 Model 15a
 Model 16b
Black Women (n = 315) Black Men (n = 286) White Women (n = 323) White Men (n = 285)
Personal UTBP 0.55 0.03 0.68 0.35 -0.92 -1.67 0.37 0.39
(1.70) (1.87) (0.97) (1.09) (2.05) (2.26) (1.22) (1.36)
Vicarious UTBP c 2.13 1.99 2.24† 1.84 2.25 1.87 2.95† 2.91
(1.33) (1.39) (1.26) (1.36) (1.65) (1.73) (1.72) (1.84)
Adjusted R2 -0.02 -0.04 0.06 0.06 0.21 0.23 0.05 0.02
Open in a new tab

† p <.10; * p <. 05; ** p <. 01; *** p< .001 (two-tailed tests); standard errors in parentheses. UTBP = unfair treatment by police.

a All models include covariates for age, gender, parental education, and financial strain in childhood.

b Includes all study covariates.

c The reference group consists of those that did not report any type of unfair treatment by police.

References

  1. Adam T.C., Epel E.S. Stress, eating and the reward system. Physiology Behavior. 2007;91(4):449–458. doi: 10.1016/j.physbeh.2007.04.011. [DOI] [PubMed] [Google Scholar]
  2. Adler N.E., Boyce W.T., Chesney M.A., Folkman S., Syme S.L. Socioeconomic inequalities in health: No easy solution. JAMA. 1993;269(24):3140–3145. [PubMed] [Google Scholar]
  3. Agnew R. Experienced, vicarious, and anticipated strain: An exploratory study on physical victimization and delinquency. Justice Quarterly. 2002;19(4):603–632. [Google Scholar]
  4. Archibald P.C., Parker L., Thorpe R., Jr. Criminal justice contact, stressors, and obesity-related health problems among Black adults in the USA. Journal of Racial Ethnic Health Disparities. 2018;5(2):387–397. doi: 10.1007/s40615-017-0382-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bjornstrom E.E. Race-ethnicity, nativity, neighbourhood context and reports of unfair treatment by police. Ethnic and Racial Studies. 2015;38(12):2019–2036. [Google Scholar]
  6. Browning S.L., Cullen F.T., Cao L., Kopache R. Race and getting hassled by the police: A research note. Police Studies. 1994;17:1–12. [Google Scholar]
  7. Brunson R.K. “Police don’t like black people”: African‐American young men’s accumulated police experiences. Criminology Public Policy. 2007;6(1):71–101. [Google Scholar]
  8. Burdette A.M., Hill T.D. An examination of processes linking perceived neighborhood disorder and obesity. Soc. Sci. Med. 2008;67(1):38–46. doi: 10.1016/j.socscimed.2008.03.029. [DOI] [PubMed] [Google Scholar]
  9. Dallman M.F., Pecoraro N., la Fleur S.E. Chronic stress and comfort foods: Self-medication and abdominal obesity. Brain, Behavior, Immunity. 2005;19(4):275–280. doi: 10.1016/j.bbi.2004.11.004. [DOI] [PubMed] [Google Scholar]
  10. DeVylder J.E., Frey J.J., Cogburn C.D., Wilcox H.C., Sharpe T.L., Oh H.Y., Link B.G. Elevated prevalence of suicide attempts among victims of police violence in the USA. Journal of Urban Health. 2017:1–8. doi: 10.1007/s11524-017-0160-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. DeVylder J.E., Cogburn C., Oh H.Y., Anglin D., Smith M.E., Sharpe T., Link B. Psychotic experiences in the context of police victimization: Data from the survey of police–public encounters. Schizophrenia Bulletin. 2017;43(5):993–1001. doi: 10.1093/schbul/sbx038. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. DeVylder J.E., Oh H.Y., Nam B., Sharpe T.L., Lehmann M., Link B.G. Prevalence, demographic variation and psychological correlates of exposure to police victimisation in four US cities. Epidemiology and Psychiatric Sciences. 2016;26(5):466–477. doi: 10.1017/S2045796016000810. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Ender, P.B. (2011). Stata binary_mediation command to estimate indirect effects with a binary dependent variable and/or mediator variable. 〈http://www.ats.ucla.edu/stat/stata/ado/analysis〉. Accessed 15.09.17.
  14. Fagan J., Geller A., Davies G., West V. Street stops and broken windows revisited: Race and order maintenance policing in a safe and changing city. In: Rice S.K., White M.D., editors. Exploring race, ethnicity and policing: Essential readings. New York University Press; New York, NY: 2010. pp. 309–348. [Google Scholar]
  15. Fagan J. Recent evidence and controversies in “The New Policing”. Journal of Policy Analysis and Management. 2017;36(3):690–700. [Google Scholar]
  16. Garza, A. (2014). A herstory of the# blacklivesmatter movement. The Feminist Wire 7. 〈http://www.thefeministwire.com/2014/10/blacklivesmatter-2〉. Accessed 05 February 2018.
  17. Geller A., Fagan J., Tyler T., Link B.G. Aggressive policing and the mental health of young urban men. American Journal of Public Health. 2014;104(12):2321–2327. doi: 10.2105/AJPH.2014.302046. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Gideon’s Army (2016). “Driving while black: A report on racial profiling in metro Nashville Police Department traffic stops.” Nashville, Tennessee. 〈https://drivingwhileblacknashville.wordpress.com/gideons-army-2/〉. Accessed 01 June 2017.
  19. Glaser J. Oxford University Press; USA: 2015. Suspect race: Causes and consequences of racial profiling. [Google Scholar]
  20. Glasner T., Van der Vaart W. Applications of calendar instruments in social surveys: A review. Quality and Quantity. 2009;43:333–349. doi: 10.1007/s11135-007-9129-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Gonzalez, T. (2016). . On “Operation Safer Streets: A divide between Nashville Police and lack Lives Matter.” 〈http://nashvillepublicradio.org/post/operation-safer-streets-divide-between-nashville-police-and-black-lives-matter#stream/0〉. Accessed on 21 August 2017.
  22. Harris D.A. The stories, the statistics, and the law. In: Rice S.K., White M.D., editors. Race, ethnicity, and policing: New and essential readings. New York University Press; New York, NY: 2010. pp. 36–86. [Google Scholar]
  23. Heard-Garris N.J., Cale M., Camaj L., Hamati M.C., Dominguez T.P. Transmitting trauma: A systematic review of vicarious racism and child health. Social Science Medicine. 2017;199:230–240. doi: 10.1016/j.socscimed.2017.04.018. [DOI] [PubMed] [Google Scholar]
  24. Henning K. Boys to men: The role of policing in the socialization of black boys. In: Davis A.J., editor. Policing the Black man. Pantheon Books; New York, NY: 2017. pp. 57–94. [Google Scholar]
  25. Heymann P.B. The new policing. Fordham Urban Law Journal. 2000;28:407–456. [Google Scholar]
  26. Hicken M.T., Lee H., Hing A.K. The weight of racism: Vigilance and racial inequalities in weight-related measures. Social Science Medicine. 2017;199:157–166. doi: 10.1016/j.socscimed.2017.03.058. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Hummer R.A. Black-white differences in health and mortality: A review and conceptual model. Soc. Q. 1996;37(1):105–125. [Google Scholar]
  28. Jackson C.L., Szklo M., Yeh H.C., Wang N.Y., Dray-Spira R., Thorpe R., Brancati F.L. Black-white disparities in overweight and obesity trends by educational attainment in the United States, 1997–2008. Journal of Obesity. 2013;2013:140743. doi: 10.1155/2013/140743. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Kelling, G. & Bratton, W.J. (2015). 〈https://www.city-journal.org/html/why-we-need-broken-windows-policing-13696.html〉. Accessed 02.02.18.
  30. Kelling G., Cole G. The Free Press; New York: 1996. Fixing broken windows. [Google Scholar]
  31. Krueger P.M., Reither E.N. Mind the gap: Race/ethnic and socioeconomic disparities in obesity. Current Diabetes Reports. 2015;15(11):95. doi: 10.1007/s11892-015-0666-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Landerman L.R., Land K.C., Pieper C.F. An empirical evaluation of the predictive mean matching method for imputing missing values. Sociological Methods Research. 1997;26(1):3–33. [Google Scholar]
  33. MacKinnon D.P. Lawrence Erlbaum Associates; Mahwah, NJ: 2008. Introduction to statistical mediation analysis. [Google Scholar]
  34. Mensah G.A., Mokdad A.H., Ford E.S., Greenlund K.J., Croft J.B. State of disparities in cardiovascular health in the United States. Circulation. 2005;111(10):1233–1241. doi: 10.1161/01.CIR.0000158136.76824.04. [DOI] [PubMed] [Google Scholar]
  35. McFarland M.J., Taylor J., Hill T.D., Friedman K.L. Stressful life events in early life and leukocyte telomere length in adulthood. Advances in Life Course Research. 2018;35:37–45. [Google Scholar]
  36. Patel V.C., Spaeth A.M., Basner M. Relationships between time use and obesity in a representative sample of Americans. Obesity. 2016;24(10):2164–2175. doi: 10.1002/oby.21596. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Priest N., Paradies Y., Trenerry B., Truong M., Karlsen S., Kelly Y. A systematic review of studies examining the relationship between reported racism and health and wellbeing for children and young people. Social Science Medicine. 2013;95:115–127. doi: 10.1016/j.socscimed.2012.11.031. [DOI] [PubMed] [Google Scholar]
  38. Ogden C.L., Carroll M.D., Kit B.K., Flegal K.M. Prevalence of childhood and adult obesity in the United States, 2011–2012. Journal of the American Medical Association. 2014;311(8):806–814. doi: 10.1001/jama.2014.732. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Oh H., DeVylder J., Hunt G. Effect of police training and accountability on the mental health of African American Adults. American Journal of Public Health. 2017;107(10):1588–1590. doi: 10.2105/AJPH.2017.304012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Rosenbaum D.P., Schuck A.M., Costello S.K., Hawkins D.F., Ring M.K. Attitudes toward the police: The effects of direct and vicarious experience. Police Quarterly. 2005;8(3):343–365. [Google Scholar]
  41. The Robert Wood Johnson Foundation, National Public Radio, Harvard Chan School of Public Health 2017. Discrimination in America. https://www.rwjf.org/en/library/research/2017/10/discrimination-in-america--experiences-and-views.html. Accessed February 2, 2018.
  42. Sewell A.A. The illness associations of police violence: Differential relationships by ethnoracial composition. Sociological Forum. 2017;32(S1):975–997. [Google Scholar]
  43. Sewell A.A., Jefferson K.A. Collateral damage: The health effects of invasive police encounters in New York City. Journal of Urban Health. 2016;93(1):42–67. doi: 10.1007/s11524-015-0016-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Sewell A.A., Jefferson K.A., Lee H. Living under surveillance: Gender, psychological distress, and stop-question-and-frisk policing in New York City. Social Science Medicine. 2016;159:1–13. doi: 10.1016/j.socscimed.2016.04.024. [DOI] [PubMed] [Google Scholar]
  45. Sternthal M.J., Slopen N., Williams D.R. Racial disparities in health: How much does stress really matter? Du Bois Review: Social Science Research on Race. 2011;8(1):95–113. doi: 10.1017/S1742058X11000087. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Sugie N.F., Turney K. Beyond incarceration: criminal justice contact and mental health. Am. Soc. Rev. 2017;82(4):719–743. [Google Scholar]
  47. Thomas, R. (2013). ‘The Talk’ has a double meaning for black people. Huffington Post. 〈http://www.huffingtonpost.com/ron-thomas/the-talk-has-a-double-mea_b_3645809.html〉. Accessed on August 21, 2017.
  48. Turner R.J. Understanding health disparities: The relevance of the stress process model. Society and Mental Health. 2012;3(3):170–186. [Google Scholar]
  49. Turner R.J., Lloyd D.A. The stress process and the social distribution of depression. Journal of Health and Social Behavior. 1999;40:374–404. [PubMed] [Google Scholar]
  50. Tyler T.R., Wakslak C.J. Profiling and police legitimacy: Procedural justice, attributions of motive, and acceptance of police authority. Criminology. 2004;42(2):253–282. [Google Scholar]
  51. Wang Y., Beydoun M.A. The obesity epidemic in the United States—gender, age, socioeconomic, racial/ethnic, and geographic characteristics: A systematic review and meta-regression analysis. Epidemiologic Reviews. 2007;29(1):6–28. doi: 10.1093/epirev/mxm007. [DOI] [PubMed] [Google Scholar]
  52. Warren P.Y. Perceptions of police disrespect during vehicle stops: A race-based analysis. Crime Delinquency. 2011;57(3):356–376. [Google Scholar]
  53. Weitzer R., Tuch S.A. Racially biased policing: Determinants of citizen perceptions. Social Forces. 2005;83(3):1009–1030. [Google Scholar]
  54. Williams D.R., Mohammed S.A. Racism and health I: Pathways and scientific evidence. American Behavioral Scientist. 2013;57(8):1152–1173. doi: 10.1177/0002764213487340. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Williams D.R., Yu Y., Jackson J.S., Anderson N.B. Racial differences in physical and mental health: Socio-economic status, stress and discrimination. Journal of Health Psychology. 1997;2(3):335–351. doi: 10.1177/135910539700200305. [DOI] [PubMed] [Google Scholar]
  56. Whitaker T.R., Snell C.L. Parenting while powerless: Consequences of “the talk”. Journal of Human Behavior in the Social Environment. 2016;26:303–309. [Google Scholar]

Articles from SSM - Population Health are provided here courtesy of Elsevier

RESOURCES